感測器網路無線通道上相關高斯雜訊之頻道等化

Abstract

有鑑於設置在惡劣環境中的無線感測器的高故障率對於無線感測器網路的系統偵測正確率有極大的影響，最近已提出運用錯誤更正碼技術之分散式分類資訊整合方法 (Distributed Classification Fusion using Error Correcting Codes or DCFECC) 作為解決方式。這個方法可在有限的電池能量支援下，提供相當高的系統容錯能力。之前有關運用錯誤更正碼技術之分散式分類資訊整合方法的研究，都是在無線通道雜訊互相獨立且相同分佈 (independent and identically distributed) 的假設下進行。在這篇論文中，我們將此假設進一步延伸，討論在空間雜訊相關的情況下，運用錯誤更正碼技術之分散式分類資訊整合方法的實現可行性。簡言之，我們的系統包含通道估計及等化，與使用最小歐式距離整合規則 (Minimum-Euclidean Distance fusion rule or MED fusion rule)。實驗顯示，此規則在空間相關的無線通道雜訊環境，確實具有不錯的容錯能力。另外，我們也提供了一個簡單的碼搜尋準則，可在低運算複雜度之下，完成碼矩陣的設計。模擬的結果顯示，最小歐式距離整合規則在白色高斯雜訊通道、空間相關雜訊通道、以及互相獨立但不同分佈的雜訊通道當中，都展現對抗故障感測器的高容錯能力。

Distributed classification fusion using error correcting codes(DCFECC) has recently been proposed for wireless sensor networks operating in a harsh environment. It has been shown to provide a considerable fault-tolerance capability against unexpected sensor faults under limited energy support. In this thesis, we extend the DCFECC approach by relaxing the assumption of independently and identically distributed wireless link noises to correlated ones. Through channel estimation and equalization, we obtain a fault-tolerant minimum Euclidean distance (MED) fusion rule suitable for use under correlated wireless link noises. A simple code search criterion is also proposed to make the code matrix design feasible with acceptable computational complexity. Simulation results show that the proposed MED fusion rule truly achieves the desired robustness against sensor faults under the simulated AWGN channels, spatially correlated channels and non-identical uncorrelated channels.